BLOOD FLOW QUANTIFICATION IN THE AORTOILIAC ARTERIES - FROM BENCH TO BEDSIDE
The PhD defence of Stefan Engelhard will take place (partly) online.
The PhD defence can be followed by a live stream.
Stefan Engelhard is a PhD student in the research group Physics of Fluids (POF). Supervisors are prof.dr. M. Versluis and prof.dr. M.M.P.J. Reijnen from the Faculty of Science & Technology (S&T).
Although the pathologic mechanisms behind aortoiliac occlusive disease (AIOD) are strongly influenced by local blood flow patterns, these are not important factors in current clinical pratice for the diagnosis and treatment of the disease. However, accurate quantification of these blood flow patterns could improve the diagnosis and treatment of patients with AIOD. Unfortunately, real-time blood flow quantification is extremely challenging.
A promising technique for this application is echoPIV, that combines high-frame-rate contrast-enhanced US (HFR-CEUS) with particle image velocimetry (PIV). The main goal of this thesis was to investigate the feasibility and clinical application of echoPIV, in order to facilitate the translation of this technique from the bench to the patient’s bedside.
In chapter 2, the link between local blood flow patterns and AIOD is described in detail. Blood flow patterns influence the development and progression of atherosclerosis and could therefore play an important role in the clinical monitoring of the disease. Furthermore, local blood flow influences the patency of stents and detailed information on flow disturbances could be used to improve endovascular treatment and post- operative care.
Current diagnostic techniques, such as conventional Duplex Ultrasound and CT- angiography, are also discussed in chapter 2, including their limitations in the evaluation of local hemodynamic factors. Here, it has become clear that there is a need for an improved modality to estimate lesion severity and to predict disease progression and stent patency, based on measured local blood flow parameters. This chapter then provides an extensive overview of novel blood flow quantification techniques, including ultrasound-based methods such as echoPIV, 4D flow MRI, and computational fluid dynamics (CFD).
Novel techniques, that allow detailed (and real-time) blood flow quantification in vivo, are not yet widely available in daily clinical practice and therefore cannot be used to guide patient-specific treatment decisions. However, valuable general insights into blood flow related issues can already be obtained in vitro. In chapter 3, local blood flow patterns were quantified using laserPIV, in two simplified models of the aortic bifurcation, with and without a Covered Endovascular Reconstruction of the Aortic Bifurcation (CERAB). The goal of this study was to investigate a common clinical problem, the re-occlusion of a stent due to compromised distal outflow.
Our study showed that local wall shear stress (WSS) decreases more in the CERAB configuration, compared to the control model, when a distal stenosis is added to the flow setup. These findings could explain the relation between distal run-off and stent re- occlusion, which is valuable insight for the development of novel endovascular devices and treatment protocols.
The dot on the horizon, however, is accurate quantification of blood flow in patients. To introduce new flow quantification techniques into the hospital, the feasibility of these techniques must be evaluated and their added value must be demonstrated. In chapters 4 and 5, the feasibility of echoPIV is investigated in healthy volunteers. The feasibility and clinical application of echoPIV in patients with aortoiliac disease, with and without stents, is investigated in chapters 6 and 7.
Chapter 4 includes a parameter study into different acquisition and processing settings. This study resulted in several important findings that were then introduced in the optimized imaging protocols for echoPIV feasibility testing in patients (chapter 6 and 7). First, microbubble destruction occurred in all but the lowest ultrasound pressure levels, which means that there is a delicate balance between contrast signal loss and a sufficiently high signal-to-noise ratio.
In Chapter 5, echoPIV-derived flow velocities from eight volunteers were compared with a reference, 4D flow MRI. Overall, the measured peak velocities during systole and the temporal profile as a whole were very similar in both techniques, indicating the validity of echoPIV. Specifically, in one volunteer that showed stagnant blood flow during diastole, both techniques showed a small and short lived recirculation at the exact same moment (end of systole) and location (origin of the left common iliac artery). This provided additional confidence in the ability of echoPIV to quantify detailed flow patterns. In two volunteers, echoPIV showed backward flow during diastole while 4D flow MRI did not. While these measurements were performed on different days, which could have explained the discrepancy, we believe that the difference is a result of the averaging of multiple heart cycles in 4D flow MRI and that real-time echoPIV displays the true velocities.
With an optimized acquisition sequence and successful validation of the echoPIV method in healthy volunteers, the technique was ready for application in patients. In Chapter 6, the feasibility and clinical application of echoPIV in patients with aortoiliac disease was investigated. This study shows that flow quantification with echoPIV was feasible in 99% of the patients, although some limiting issues occur in over half of the patients leading to an incomplete visualization, i.e. only during part of the cardiac cycle, or only in parts of the imaged vessel segment. Despite these limitations, echoPIV-derived flow parameters (vector complexity and vorticity) could be used to successfully identify blood flow disturbances in regions with normal peak systolic velocity, that would not be identified as problematic using conventional duplex US.
In chapter 7, the robustness of echoPIV is investigated, by testing the feasibility of flow quantification inside stents placed in the aortoiliac region. This study shows that echoPIV is feasible near and inside a wide variety of commonly used stents. Furthermore, these stents do not degrade the image quality or tracking performance of the echoPIV algorithm. On the contrary, in 4D flow MRI a variety stents interfere with the magnetic field, resulting in imaging artefacts and inaccurate flow quantification. This demonstrated benefit of echoPIV, together with the cost-effectiveness and ease-of-use of ultrasound, shows the potential for implementation into a clinical workflow, despite the need for administration of a contrast agent. This could drastically improve the future outcomes of endovascular treatment in the aortoiliac tract.
